**** BEGIN LOGGING AT Mon Jul 18 02:59:58 2011
Jul 18 12:55:00 <ant_work> Jay7: iirc the initramfs still needs to be prepopulated by a micro-devices-table, even with devtmpfs
Jul 18 12:55:12 <ant_work> isn't?
Jul 18 21:15:19 <ant__> Jay7: hello
Jul 18 21:15:30 <Jay7> hi ant__
Jul 18 21:15:42 <Jay7> about devtmpfs - yes
Jul 18 21:15:57 <Jay7> devices are not appears automagically..
Jul 18 21:16:33 <ant__> koen suggested to drop pre-2.6.32
Jul 18 21:17:23 <ant__> well, you could do a release 0.9 and then require 2.6.32 from 1.0
Jul 18 21:17:41 <ant__> or apply that devtmpfs patch
Jul 18 21:18:27 <Jay7> hm.. why do we need devtmpfs?
Jul 18 21:18:48 <ant__> I asked to extend device_tables-minimal in oe-core
Jul 18 21:19:02 <ant__> koen replied'why don't use devtmpfs?'
Jul 18 21:19:20 <Jay7> because devtmpfs does not populate devices w/o devices manager
Jul 18 21:19:29 <Jay7> or I misunderstand things
Jul 18 21:19:54 <ant__> once the kernel starts devtmps can do the basics
Jul 18 21:20:02 <Jay7> but iirc we have had no devices with devtmpfs but w/o udev
Jul 18 21:20:07 <ant__> for the kernel to boot you need the basic devices...
Jul 18 21:21:06 <Jay7> hm..
Jul 18 21:21:11 <Jay7> but it should
Jul 18 21:21:11 <ant__> no, that is not the problem
Jul 18 21:21:19 <ant__> the issue was the delay
Jul 18 21:22:06 <Jay7> Device Drivers ---> Generic Driver Options --->
Jul 18 21:22:08 <Jay7> [*] Create a kernel maintained /dev tmpfs
Jul 18 21:22:09 <Jay7> [*] Automount devtmpfs at /dev
Jul 18 21:22:14 <Jay7> we need this in kernel then
Jul 18 21:23:08 <ant__> yes, is default and takes 20kb compressed
Jul 18 21:23:36 <ant__> is disabled now
Jul 18 21:27:40 <Jay7> well.. I can import devtmpfs patch again
Jul 18 21:27:45 <ant__> Jay7: btw...you could reply and say we have code for device recreation (link)
Jul 18 21:28:03 <Jay7> and we may just use --enable-devtmpfs/--disable-devtmpfs
Jul 18 21:28:20 <ant__> so we make oe-core aware of the project :)
Jul 18 21:28:31 <Jay7> but we need /dev/null, /dev/tty*, /dev/console, etc
Jul 18 21:28:55 <Jay7> and /dev/input/event*
Jul 18 21:29:00 <Jay7> or /dev/event*
Jul 18 21:30:13 <ant__> yes, device_table-micro.txt
Jul 18 21:30:54 <ant__> http://paste.debian.net/123283/
Jul 18 21:31:03 <ant__> iirc I used this one during tests
Jul 18 21:31:22 <Jay7> no event*
Jul 18 21:31:46 <Jay7> well.. just enable devtmpfs and try again
Jul 18 21:32:04 <ant__> well, better here
Jul 18 21:32:05 <ant__> http://paste.debian.net/123284/
Jul 18 21:32:09 <Jay7> with "[*] Automount devtmpfs at /dev" it should work event w/o patch
Jul 18 21:32:22 <ant__> no
Jul 18 21:32:31 <ant__> not valid in case of initramfs
Jul 18 21:32:37 <ant__> must be done manually
Jul 18 21:32:46 <ant__> (it was so last time I checked...)
Jul 18 21:33:54 <ant__> see help text
Jul 18 21:35:31 <ant__> http://lxr.free-electrons.com/source/drivers/base/Kconfig
Jul 18 21:35:34 <ant__> line 47
Jul 18 21:36:28 <Jay7> +	  This option does not affect initramfs based booting, here
Jul 18 21:36:30 <Jay7> +	  the devtmpfs filesystem always needs to be mouted manually
Jul 18 21:36:31 <Jay7> +	  after the roots is mounted.
Jul 18 21:36:42 <Jay7> well.. then we need patch
Jul 18 21:37:16 <ant__> last lines are interesting...
Jul 18 21:37:24 <ant__> ..completely empty...
Jul 18 21:38:11 <ant__> but I don't think for initramfs
Jul 18 21:38:23 <ant__> how do you read it?
Jul 18 21:39:18 <Jay7> same - initramfs have exception
Jul 18 21:39:35 <Jay7> i.e. you should mount it manually in initramfs-based boot
Jul 18 21:40:09 <ant__> yes, and the basic nodes are required
Jul 18 21:40:27 <ant__> we can try again, I don't remember much
Jul 18 21:40:47 <ant__> and yes, size of /dev was less than 20kb :)
Jul 18 21:41:14 <ant__> I remember we stopped testing, no benefits
Jul 18 21:42:28 <Jay7> imho we don't needed device-table then
Jul 18 21:50:34 <Jay7> well.. I'll reintegrate devtmpfs patch then
Jul 18 21:50:54 <Jay7> now sleep
Jul 18 21:58:10 <ant__> gn
**** ENDING LOGGING AT Tue Jul 19 02:59:56 2011